The long-term goal of this research is to understand how chemokines recognize their binding partners in order to develop molecules that alter chemokine signaling for therapeutic benefit. Our previous work defined the molecular details and functional importance of self-association, glycosaminoglycan binding, and receptor sulfotyrosine recognition for the chemokine CXCL12 and its receptor CXCR4. We now propose structure- function studies of intact chemokine-receptor complexes while translating our knowledge of CXCR4 biased agonism into clinical applications of high significance, with an initial focus on aggressive or refractory cancers. Chemokine signaling is initiated through a two-site, two-step process involving distinct interactions with the extracellular N-terminal domain of the GPCR (site 1) and the transmembrane domain (site 2). In the previous funding period we solved the first structure of a 'site 1' complex for a CXCL12 dimer bound to the CXCR4 N- terminus and unexpectedly found that the CXCL12 dimer is a partial CXCR4 agonist that potently inhibits cell migration while stimulating G protein signaling and intracellular calcium release. This was the first demonstration that chemokine receptors are capable of agonist-biased signaling, a recently discovered phenomenon in which multiple ligands for a single GPCR can elicit distinct functional states and activate different combinations of downstream signaling events. The focus of our studies in this renewal application is the chemokine CXCL12, which binds the receptors CXCR4 and CXCR7, and CXCL11, a chemokine that binds CXCR7 but not CXCR4. In aim 1, we will test the hypothesis that agonist biased CXCR4 signaling arises from structurally distinct ligand bound states of the receptor. In aim 2, we will translate our knowledge of structure- function relationships for the CXCL12-CXCR4 axis into a novel therapeutic strategy for metastatic cancer. In the previous funding period, we developed and patented an engineered CXCL12 dimer protein that functions as a partial CXCR4 agonist and potent inhibitor of cancer progression in multiple animal models. We will optimize the properties of this molecule for in vivo use and initiate pre-clinical trials using state-of-the-art animal modls for pancreatic cancer. Finally, to understand chemokine-receptor selectivity, we will solve and compare NMR structures of soluble 'site 1' complexes involving promiscuous ligands and receptors (aim 3).